CN116362394A - Synergistic prediction method and system for marine algae growth pollution - Google Patents

Abstract

The invention discloses a co-integration prediction method and system for marine algae growth pollution, belonging to the technical field of algae pollution control. According to the existing empirical theory of algae growth factors and relevant results of nonlinear complex system theory, the present invention performs statistical analysis and multiple linear regression prediction based on real data related to algae growth, obtains the core factors affecting algae growth, and then performs marine algae pollution. governance. According to the actual conditions of different sea areas, the influencing factors on the growth of marine algae can be obtained, and the pollution of marine algae can be controlled in a targeted manner; the problem of "lack of effective methods for predicting the influencing factors of marine algae pollution" in the prior art is solved.

Description

A cointegration prediction method and system for marine algae growth pollution

technical field

The invention relates to the technical field of algae pollution control, in particular to a cointegration prediction method and system for marine algae growth pollution.

Background technique

The statements in this section merely mention the background technology related to the present invention and do not necessarily constitute the prior art.

Under special environmental conditions, the excessive reproduction and high accumulation of certain marine algae (ie "harmful algal blooms") will cause marine algae pollution. At present, the methods for judging marine algae pollution include traditional ship regular surveys and shoreside artificial regular observations, but it is difficult to detect sudden changes in red tides; or use multi-source satellite remote sensing, drones, ships, and vehicles to monitor green tides three-dimensionally Using ecological dynamics and drift prediction models to predict changes in green tides, and the long-term equilibrium and causal relationship between green tide coverage area and sea temperature to predict the scale changes of green tides.

All of the above marine algae pollution predictions require an in-depth understanding of algae growth quantity or density changes. When the segmental and vertical distribution changes, and when the climatic factors, wind speed and temperature are different, there will be different growth and change characteristics. In recent years, due to human fishing, sewage discharge, eutrophication of water bodies, and changes in natural climate, etc., the ecological balance of the water ecosystem has been destroyed, resulting in the decline of trophic levels and the decline of dominant species.

Therefore, even if people have a general understanding of the increase in fertilizer use and global warming as possible reasons for the deterioration of the algal bloom situation, there is still no effective way to control marine algae pollution. Although there are algorithms such as machine learning to classify the influencing factors of algae growth, the machine learning algorithm does not analyze the influencing factors and cannot provide help for the pollution control of marine algae.

Contents of the invention

In order to solve the deficiencies of the prior art, the present invention provides a cointegration prediction method, system, electronic equipment and computer-readable storage medium for marine algae growth pollution, based on the existing empirical theory of algae growth factors and the theory of nonlinear complex systems The relevant results are based on real data related to algae growth in different regions, and statistical analysis and multiple linear regression predictions are carried out.

In the first aspect, the present invention provides a cointegration prediction method for marine algae growth pollution;

A cointegration prediction method for marine algae growth pollution, comprising:

Obtain the data related to the growth of marine algae and preprocess the data related to the growth of marine algae;

Input the pre-processed marine algae growth-related data into the preset co-integration prediction model to obtain the co-integration relationship between the variables affecting the growth of marine algae, so as to inhibit the pollution of marine algae growth according to the co-integration relationship; The processed marine algae growth-related data are input into the preset co-integration prediction model for processing including:

Perform stability analysis on the preprocessed data related to the growth of marine algae, and obtain the multivariate temporal dynamic growth trend curve of algae;

Normalize the data related to the growth of marine algae after preprocessing, and determine the key influencing variables and secondary influencing variables of algae growth pollution based on the multivariate time dynamic growth trend curve of algae;

Calculate the correlation coefficient matrix between the key influencing variables and the secondary influencing variables, and establish the cointegration relationship between the key influencing variables and the secondary influencing variables; use the Jcitest method to test the cointegration relationship between the key influencing variables and the secondary influencing variables , to build a multiple linear regression model. Further, the stability analysis is carried out on the data related to the growth of marine algae after pretreatment, and the multivariate time dynamic growth trend curve of the obtained algae is specifically:

According to the data related to the growth of marine algae after preprocessing, the number of occurrences of marine algae growth pollution or the data related to the growth of marine algae growth showing a decreasing trend is analyzed to obtain a multivariate time dynamic growth trend curve.

Further, the determination of key influencing variables and secondary influencing variables of algae growth pollution based on the multivariate time dynamic growth trend curve of algae includes:

Based on the multi-time dynamic growth trend curve of algae, the chlorophyll-a concentration of algae is regarded as the key influencing variable of algae growth pollution;

Correlation analysis is carried out on the basic data of marine water quality in the data related to the growth of marine algae to obtain secondary influencing variables with high correlation.

Preferably, the secondary influencing variables with low correlation are deleted, and the Dickey Fuller test method is used to analyze the stationarity of the processed data;

The QQ graph analysis method was used to judge whether the core influencing variables conformed to the normal distribution.

Further, the multiple linear regression model is expressed as

Y＝1.63X ₁ +4.6516X ₂ +3.6104X ₃ -0.91236X ₄

+1.0207X ₅ -2.563X ₆ +0.52605X ₇ +0.33509X ₈

+0.85703X ₉ -0.023949

Among them, Y is the concentration of chlorophyll a, X ₁ is the five-day biochemical oxygen demand, X ₂ is the content of ammonia nitrogen, X ₃ is the temperature, X ₄ is the salinity, X ₅ is the content of pheomagnesium pigment, X ₆ is the content of silicon, X ₇ is dissolved oxygen content, X ₈ is total nitrogen content, X ₉ is total phosphorus content;

or,

The multiple linear regression model is expressed as

Y＝-0.023616X ₁ +1.4527X ₂ -0.29186X ₃ +0.055603X ₄ +0.13686

Among them, Y is the ratio of silicon to phosphorus, _X1 is the average concentration of chlorophyll a in the water column, _X2 is silicate, _X3 is dissolved inorganic nitrogen, and _X4 is the ratio of nitrogen to phosphorus.

Further, preprocessing the data related to the growth of marine algae includes:

Construct a data set related to marine algae growth, and determine whether the missing values in the data set affect the prediction of marine algae growth pollution;

If so, the missing value is supplemented by interpolation; if not, the missing value is deleted;

Non-numerical data not relevant to marine algae growth pollution predictions were removed.

Further, the data related to the growth of marine algae includes chlorophyll concentration, sea water nutrient concentration and sea water quality data.

In the second aspect, the present invention provides a cointegration prediction system for marine algae growth pollution;

A cointegration prediction system for marine algae growth pollution, including:

The data processing module is configured to: obtain data related to the growth of marine algae, and preprocess the data related to the growth of marine algae;

The co-integration relationship prediction module is configured to: input the pre-processed marine algae growth-related data into the preset co-integration prediction model for processing, and obtain the co-integration relationship between the variables affecting the growth of marine algae, so as to suppress according to the co-integration relationship Marine algae growth pollution; wherein, inputting the preprocessed marine algae growth-related data into a preset co-integration prediction model for processing includes:

Perform stability analysis on the preprocessed data related to the growth of marine algae, and obtain the multivariate temporal dynamic growth trend curve of algae;

Normalize the data related to the growth of marine algae after preprocessing, and determine the key influencing variables and secondary influencing variables of algae growth pollution based on the multivariate time dynamic growth trend curve of algae;

Calculate the correlation coefficient matrix between the key influencing variables and the secondary influencing variables, and establish the cointegration relationship between the key influencing variables and the secondary influencing variables; use the Jcitest method to test the cointegration relationship between the key influencing variables and the secondary influencing variables , to establish a multiple linear regression model to adjust the proportion of secondary impact variables according to the co-integration relationship, and to inhibit the growth and pollution of marine algae.

In a third aspect, the present invention provides an electronic device;

An electronic device includes a memory, a processor, and computer instructions stored in the memory and run on the processor. When the computer instructions are run by the processor, the steps of the above-mentioned cointegration prediction method for marine algal growth pollution are completed.

In a fourth aspect, the present invention provides a computer-readable storage medium;

A computer-readable storage medium is used for storing computer instructions, and when the computer instructions are executed by a processor, the steps of the above-mentioned cointegration prediction method for marine algae growth pollution are completed.

Compared with prior art, the beneficial effect of the present invention is:

The technical solution provided by the present invention is based on the complex nonlinear dynamic characteristics of algae growth, the characteristics of actual algae growth data, and in combination with the regional conditions of the economic development level, constructs a time dynamic comparison multiple linear regression model based on real algae growth related data, which is helpful Find out the relationship between human activities and algae pollution, and determine the key factors that affect the growth of marine algae pollution, so as to facilitate regulation based on key factors, inhibit the growth of marine algae, and remediate the growth of marine algae pollution.

Description of drawings

The accompanying drawings constituting a part of the present invention are used to provide a further understanding of the present invention, and the schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations to the present invention.

Fig. 1 is a schematic flow diagram provided by an embodiment of the present invention;

Fig. 2 is a schematic diagram of the multi-time dynamic growth trend curve of algae in a certain sea area with a general economic development level provided by the embodiment of the present invention; wherein, (a) is a schematic diagram of the time dynamic growth trend curve of water body nutrients, and (b) is a graph of chlorophyll a Schematic diagram of time dynamic growth trend curve;

Fig. 3 is a schematic diagram of the correlation relationship between the silicon-phosphorus ratio and other relevant data in a certain sea area with a general level of economic development provided by the embodiment of the present invention;

Fig. 4 is a schematic diagram of co-integration relationship test between the silicon-phosphorus ratio and other related data in a certain sea area with general economic development level provided by the embodiment of the present invention; wherein, (a) is the scatter plot and regression confidence interval of the predicted value and the actual value Figure, (b) is the normality distribution test graph of predicted value and true value residual, (c) is the scatter diagram of fitting value and residual value relation;

Fig. 5 is a schematic diagram of the multi-time dynamic growth trend curve of algae in a certain sea area with a high level of economic development provided by the embodiment of the present invention; wherein, (a) is a schematic diagram of the time dynamic growth trend curve of chlorophyll a, and (b) is the water body nutrient salt Schematic diagram of time dynamic growth trend curve;

Fig. 6 is a schematic diagram of the correlation coefficient between variables such as water body nutrients and chlorophyll a in a certain sea area with a high level of economic development provided by the embodiment of the present invention;

Figure 7 is a schematic diagram of the co-integration relationship test between chlorophyll a and secondary influencing variables in a certain sea area with a high level of economic development provided by the embodiment of the present invention; wherein, (a) is the true chlorophyll a concentration and the fitted chlorophyll a Concentration map, (b) is a residual map, (c) is a schematic diagram of the probability distribution of the residual map; (d) is a scatter plot and a regression confidence interval map of model predicted values and existing statistical data values.

Detailed ways

It should be noted that the following detailed description is exemplary and intended to provide further explanation of the present invention. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

It should be noted that the terminology used here is only for describing specific embodiments, and is not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly dictates otherwise, the singular is intended to include the plural, and it should also be understood that the terms "comprising" and "having" and any variations thereof are intended to cover a non-exclusive Comprising, for example, a process, method, system, product, or device comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include steps or units not explicitly listed or for these processes, methods, Other steps or units inherent in a product or equipment.

In the case of no conflict, the embodiments and the features in the embodiments of the present invention can be combined with each other.

Embodiment one

Combined with Figure 1, the existing machine learning algorithms do not pay attention to the correlation and causality behind the factors affecting algae growth. Therefore, they cannot give the quantitative coefficients of different factors or factor combinations (ie, multivariate cointegration) affecting the concentration of algae growth. . Among them, the co-integration relationship refers to a long-term stable relationship presented by the time series starting from the actual data generation process; only when there is a co-integration relationship in the time series can regression analysis and prediction be performed. If the time series is not stable and the regression analysis is performed directly, the significance and fitting degree of the regression results will be very good, but such a regression relationship does not actually exist. Therefore, the present invention provides a method for predicting cointegration of marine algae growth pollution, which includes the following steps:

S1. Obtain the data related to the growth of marine algae, and preprocess the data related to the growth of marine algae; specifically, construct a data set related to the growth of marine algae, and determine whether the missing values in the data set affect the cointegration prediction of marine algae growth pollution; if so, then Missing values were supplemented by interpolation; if not, missing values were deleted; non-numeric data irrelevant to cointegration prediction of marine algae growth pollution were deleted.

S2. Input the pre-processed data related to the growth of marine algae into the preset co-integration prediction model for processing, and obtain the co-integration relationship between the variables affecting the growth of marine algae, so as to inhibit the growth and pollution of marine algae according to the co-integration relationship; wherein, the The preprocessed marine algae growth-related data is input into the preset co-integration prediction model for processing including:

S201. Perform stability analysis on the preprocessed data related to the growth of marine algae, and obtain the multivariate time dynamic growth trend curve of the algae; specifically, according to the data related to the growth of the preprocessed marine algae, analyze the occurrence times of marine algae growth pollution or marine algae The data related to the growth of marine algae whose polluted area shows a decreasing trend is obtained, and the multivariate time dynamic growth trend curve is obtained.

Based on the pattern dynamics in the theory of nonlinear dynamical systems, it can be known that the spatial or dynamic changes in the number of marine algae individuals are related to behaviors such as diffusion and migration. They are open and dissipative systems far from thermodynamic equilibrium. Under the effects of capture and toxins It presents complex features, and the obvious feature of a complex system is that from a spatial point of view, local laws cannot predict overall changes; from a time point of view, short-term laws cannot predict long-term changes well. Therefore, it is not that the more data the better, but it must be based on the real algae growth data in different sea areas and analyze its specific dynamic trends in order to give a better forecast in a short period of time.

Specifically, according to the data related to the growth of marine algae after preprocessing, logarithmize the data, and analyze its stability, and then give the relevant data of the growth of marine algae that shows the number of occurrences of marine algae growth pollution or the area of marine algae growth pollution that shows a decreasing trend , to obtain the multivariate time dynamic growth trend curve.

S202. Perform normalization processing on the preprocessed data related to the growth of marine algae, and determine the key influencing variables and secondary influencing variables of algae growth pollution based on the multivariate time dynamic growth trend curve of algae. Including the following steps:

S2021. Based on the multivariate time dynamic growth trend curve of algae, obtain the key influencing variables of algae growth pollution;

S2022. Carry out correlation analysis on the basic data of marine water quality in the data related to the growth of marine algae, and obtain highly correlated secondary influencing variables;

S2023. Delete secondary influencing variables with low correlation, and use the Dickey Fuller test method to analyze the stationarity of the processed data; the specific process is as follows:

(1) Assume that the tested time series obeys the autoregressive model:

Y _t ＝δY _t-1 +ε _t ,

(2) Analyze whether the λ in its differential regression model is 0:

ΔY _t =λY _t-1 +ε _t ,

(3) If it is 0, the tested time series is the most basic unit root process, which is non-stationary. The method for judging the significance level is the Monte Carlo simulation method.

S203. Calculate the correlation coefficient matrix between the key influencing variable and the secondary influencing variable, which is used to select the independent variable that affects the predicted variable. At the same time, based on the definition of cointegration, finally analyze the influencing factors with cointegration relationship, and obtain multivariate For the specific coefficients and statistical indicators of the linear regression model, use the Jcitest method to test the co-integration relationship between the key influencing variables and the secondary influencing variables, and establish a multiple linear regression model; use the QQ graph analysis method to judge whether the residuals of the multiple linear regression pre-model are Fits a normal distribution. QQ diagram is the abbreviation of Quantile-Quantile (quantile-quantile diagram). Its principle is to compare the quantile of a set of sample data with the quantile of known distribution data to judge whether the test data conforms to the some kind of distribution.

Among them, the multiple linear regression model is used to display the co-integration relationship between the key influencing variables and the secondary influencing variables, so as to adjust the proportion of the secondary influencing variables according to the co-integrating relationship, and inhibit the growth and pollution of marine algae.

The normal distribution test is to judge whether the residuals of the multiple linear regression model conform to the normal distribution. This is because the model is guaranteed to be optimal only if the residuals conform to a normal distribution.

Specifically, firstly, the pairwise correlation coefficient is calculated based on the logarithmization of the original data, and the correlation coefficient matrix is obtained, which is represented by a graph.

Then, the Jcitest method is used to test the co-integration relationship between the key influencing variables and the secondary influencing variables. The Jcitest method is the Johansen cointegration test method, which is used to test whether the causal relationship between (non-stationary) data is pseudo-regression. Cointegration theory can be used to correctly explain forecasting phenomena. Jcitest is a multivariate equation technique whose idea is to use maximum likelihood estimation to test the co-integration relationship between multiple variables. The specific inspection process is as follows:

(1) Use the unit root test to determine the lagging order of the data related to the growth of marine algae;

(2) Construct a vector autoregressive (VAR) model, taking k variables with a lagging order of 1 as an example:

IID(0,σ²),Cov(u_1,t,u_2,t)＝0；Among them, Y _1,t ,Y _2,t are dependent variables, X _1,t ,X _2,t are independent variables, β _1,t, β _2,t ,π _1,t ,π _2,t are regression coefficients , the random error term u _1,t ,u _2,t satisfies u _1,t ,u _2,t

IID(0,σ ² ), Cov(u _1,t ,u _2,t )=0;

(3) Construct a Vector Error Correction (VECM) model, the residual sequence e _1,t ,e _2,t satisfy:

的秩；(4) Analyze the influence matrix in the VECM model

rank;

(401) carry out the statistical test of matrix M trace:

为按降序排列的矩阵M的特征根；in,

is the characteristic root of the matrix M arranged in descending order;

4.2. Check the number of non-zero characteristic roots of the M matrix to determine the co-integration relationship.

Further, in some embodiments, a cointegration prediction method for marine algae growth pollution disclosed in this embodiment is described in detail by taking a certain sea area with a high level of economic development as an example with reference to FIGS. 2-7 .

A cointegration prediction method for marine algae growth pollution, comprising the following steps:

Step 1. Establish a data set related to algae growth, and check whether the missing values in the data set affect the final statistical analysis results. If not, delete them; if they do, choose interpolation to supplement the missing values. For the data that does not have a specific value in the data set, but the upper or lower bound of the value is given, the critical value is selected. Based on the existing nonlinear dynamical system analysis related to algae growth and the analysis of algae actual data and field data, irrelevant non-numerical data are deleted.

The data set related to algae growth includes seawater quality data in the surface area of DM1 (data source epd.gov.hk), the time range is 1986/2/14～2021/12/4, including five-day biochemical oxygen demand (mg/L ), ammonia nitrogen (mg/L), chlorophyll a (μg/L), dissolved oxygen saturation percentage (percentage), nitrate nitrogen (mg/L), pH value, pheomagnesium pigment (μg/L), salinity, Temperature (Celsius), Inorganic Nitrogen (mg/L), Total Nitrogen (mg/L) and Total Phosphorus (mg/L).

Step 2. Using the statistical analysis results of marine algae growth nonlinear dynamic system, complex system and other theories and real data, statistically analyze the data related to algae growth in regions with developed economy but the occurrence of marine algae pollution or the pollution area shows a decreasing trend, and give its Multivariate time dynamic long-term trend.

Step 3. In order to remove the influence of different units, normalize the data with the maximum and minimum method. Based on the experimental analysis and empirical analysis results related to algae growth, chlorophyll-a in the ocean can be regarded as the output variable (core impact variable) of pollution phenomena such as marine red tide or green tide. First, based on the basic data of marine water quality in Deep Bay, analyze five-day biochemical oxygen demand (mg/L), ammonia nitrogen (mg/L), total nitrogen (mg/L), total phosphorus (mg/L), chlorophyll a (μg /L), temperature (Celsius), transparency (m), pH value, pheosin (μg/L), dissolved oxygen (mg/L), dissolved oxygen saturation percentage (%), salinity (psu), nitric acid Correlation between salt nitrogen (mg/L), nitrite nitrogen (mg/L), and orthophosphate phosphorus (μg/L), that is, the correlation between core impact variables and secondary impact variables.

According to Figure 6, it can be seen that ammonia nitrogen, total nitrogen, total phosphorus, and orthonitrate phosphorus are highly correlated, so the total nitrogen, total phosphorus, and orthonitrate phosphorus are removed, and the ammonia nitrogen data is retained to avoid highly correlated factors affecting the prediction results. Accuracy, to ensure independence between independent variables. However, because there are too many influencing factors, it is necessary to further delete irrelevant factors, and keep the five-day biochemical oxygen demand, ammonia nitrogen, salinity, pheomagnesizing pH, salinity, pheomagnesizing pigment, temperature, dissolved oxygen, total phosphorus, and total phosphorus 9 factor. And using the Dickey Fuller test (ADF) test method, it is found that the processed data set is not stable. So judge again with the first-order difference to become stationary.

Step 4, according to the multivariate data correlation coefficient matrix of algae growth-related data, statistical analysis found that there is no significant linear correlation between the stable marine algae growth-related data, X ₁ five-day biochemical oxygen demand, X ₂ ammonia nitrogen, X ₃ temperature, X ₄ salinity, X ₅ pheomegigmentation, X ₆ silicon, X ₇ dissolved oxygen, X ₈ total nitrogen, X ₉ total phosphorus 9 factors remained as the most important core variables. At the same time, according to Figure 3, it shows that the output variable Y chlorophyll a does not have a single linear correlation with all other factors, so it should be considered whether chlorophyll a has a co-integration relationship with other variables. Using Jcitest to test the co-integration relationship between the data related to the growth of marine algae, the test results found that there is indeed a co-integration relationship. Then, build a multiple linear regression model:

Y＝1.63X ₁ +4.6516X ₂ +3.6104X ₃ -0.91236X ₄

+1.0207X ₅ -2.563X ₆ +0.52605X ₇ +0.33509X ₈

+0.85703X ₉ -0.023949(1)

The estimated coefficients and corresponding statistical values of the corresponding multiple linear regression model are shown in Table 1.

Table 1 Statistics of Multiple Linear Regression

The F-statistic was 59.7 and the p-value was 7.36e-66. The above model (1) and the results of Table 1 show that there is a significant co-integration relationship between the concentration of chlorophyll a and the five-day biochemical oxygen demand, ammonia nitrogen, salinity, pheomagnesium pigment, silicon and dissolved oxygen, in which salinity, silicon The concentration of chlorophyll a played a significant negative effect, and the other four factors played a significant positive effect.

Further, in some embodiments, taking a certain sea area with a general level of economic development as an example, the cointegration prediction method for marine algae growth pollution in this embodiment is further described with reference to FIGS. 2-7 . Specific steps are as follows:

Step 1. Establish data sets related to algae growth, and check whether the missing values in the two data sets affect the final statistical analysis results. If not, delete them; if they do, choose interpolation methods to supplement missing values. For the data that does not have a specific value in the data set, but the upper or lower bound of the value is given, the critical value is selected. Based on the existing nonlinear dynamical system analysis related to algae growth and the analysis of algae actual data and field data, irrelevant non-numerical data are deleted. The data related to the growth of marine algae include the long-term data of surface chlorophyll a concentration, bottom chlorophyll a concentration, water column average chlorophyll a concentration (mg/L), and water body nutrient data such as nitrogen-phosphorus ratio, silicon-phosphorus ratio, and silicon-nitrogen ratio. , the time range is from March 1997 to November 2010.

Step 2. Using theoretical and empirical results such as nonlinear dynamical systems and complex systems of marine algae growth, statistically analyze data related to algae growth in regions with developed economies but with a decreasing trend in the number of occurrences of marine algae pollution or the polluted area, and give its multivariate time dynamic long-term trend.

Step 3, according to the correlation coefficient of algae growth-related data (see Figure 6), it is found that there is a significant correlation between the surface chlorophyll a concentration, the bottom chlorophyll a concentration and the water column average chlorophyll a concentration, the ratio of silicon to phosphorus, the ratio of silicon to nitrogen and the ratio of silicic acid Salt has a significant linear correlation, so the average chlorophyll a concentration of the retained water column, silicate, dissolved inorganic nitrogen, nitrogen-phosphorus ratio, and silicon-phosphorus ratio are core variables, and the silicon-phosphorus ratio is the output variable, and the others are core explanatory variables.

Step 4. Establish a multiple linear regression model for algae growth in a certain sea area with a general level of economic development. The multiple linear regression model for algae growth is:

Y＝-0.023616X ₁ +1.4527X ₂ -0.29186X ₃ +0.055603X ₄ +0.13686 (2)

The corresponding statistics are listed in Table 2.

The corresponding estimated coefficients and statistical values are shown in Table 2.

Table 2 Estimated coefficients and corresponding statistical values of multiple linear regression

The F-statistic is 32.6 and the p-value is 6.63e-13. The above model (2) and the results of Table 2 show that there is a significant co-integration relationship between the silicon-phosphorus ratio and the average concentration of chlorophyll a in the water column, silicate, dissolved inorganic nitrogen, and nitrogen-phosphorus ratio. The ratio has a significant negative effect, and the nitrogen-phosphorus ratio has a significant positive effect.

According to model (1) and model (2), it can be found that for the deep bay sea area with a high degree of economic development, the concentration of chlorophyll a can be significantly reduced by increasing salinity and silicon content, or reducing ammonia nitrogen content, five-day biochemical oxygen demand amount, pheomagnesin content and dissolved oxygen content to significantly reduce the concentration of chlorophyll a. For the Jiaozhou Bay area with a low level of economic development, the ratio of silicon to phosphorus can be significantly reduced by increasing the content of dissolved inorganic nitrogen, or the content of silicon and phosphorus can be significantly reduced by increasing the content of silicate and nitrogen to phosphorus ratio.

Embodiment two

This embodiment discloses a cointegration prediction system for marine algae growth pollution, including:

The data processing module is configured to: obtain data related to the growth of marine algae, and preprocess the data related to the growth of marine algae;

The co-integration relationship prediction module is configured to: input the pre-processed marine algae growth-related data into the preset co-integration prediction model for processing, and obtain the co-integration relationship between the variables affecting the growth of marine algae, so as to suppress according to the co-integration relationship Marine algae growth pollution; wherein, inputting the preprocessed marine algae growth-related data into a preset co-integration prediction model for processing includes:

Perform stability analysis on the preprocessed data related to the growth of marine algae, and obtain the multivariate temporal dynamic growth trend curve of algae;

Normalize the data related to the growth of marine algae after preprocessing, and determine the key influencing variables and secondary influencing variables of algae growth pollution based on the multivariate time dynamic growth trend curve of algae;

Calculate the correlation coefficient matrix between the key influencing variables and the secondary influencing variables, and establish the cointegration relationship between the key influencing variables and the secondary influencing variables; use the Jcitest method to test the cointegration relationship between the key influencing variables and the secondary influencing variables , to build a multiple linear regression model.

What needs to be explained here is that the above-mentioned data processing module and cointegration relationship prediction module correspond to the steps in the first embodiment, and the examples and application scenarios implemented by the above-mentioned modules and the corresponding steps are the same, but are not limited to those disclosed in the first embodiment above Content. It should be noted that, as a part of the system, the above-mentioned modules can be executed in a computer system such as a set of computer-executable instructions.

Embodiment three

Embodiment 3 of the present invention provides an electronic device, including a memory, a processor, and computer instructions stored in the memory and run on the processor. When the computer instructions are run by the processor, the above-mentioned cointegration prediction method for marine algal growth pollution is completed. step.

Embodiment four

Embodiment 4 of the present invention provides a computer-readable storage medium for storing computer instructions, and when the computer instructions are executed by a processor, the steps of the above-mentioned cointegration prediction method for marine algal growth pollution are completed.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each procedure and/or block in the flowchart and/or block diagram, and a combination of procedures and/or blocks in the flowchart and/or block diagram can be realized by computer program instructions. These computer program instructions may be provided to a general purpose computer, special purpose computer, embedded processor, or processor of other programmable data processing equipment to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing equipment produce a An apparatus for realizing the functions specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.

These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to operate in a specific manner, such that the instructions stored in the computer-readable memory produce an article of manufacture comprising instruction means, the instructions The device realizes the function specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.

These computer program instructions can also be loaded onto a computer or other programmable data processing device, and a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device Steps are provided for implementing the functions specified in the flow chart or flow charts and/or block diagram block or blocks.

The description of each embodiment in the foregoing embodiments has its own emphases, and for parts not described in detail in a certain embodiment, reference may be made to relevant descriptions of other embodiments.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims (10)

1. A cointegration prediction method for marine algae growth pollution, characterized in that, comprising: Obtain the data related to the growth of marine algae and preprocess the data related to the growth of marine algae; Input the pre-processed marine algae growth-related data into the preset co-integration prediction model for processing, and obtain the co-integration relationship between the variables affecting the growth of marine algae, so as to inhibit the pollution of marine algae growth according to the co-integration relationship; among them, the pre-processing The final data related to the growth of marine algae is input into the preset co-integration prediction model for processing including: Perform stability analysis on the preprocessed data related to the growth of marine algae, and obtain the multivariate temporal dynamic growth trend curve of algae; Normalize the data related to the growth of marine algae after preprocessing, and determine the key influencing variables and secondary influencing variables of algae growth pollution based on the multivariate time dynamic growth trend curve of algae; Calculate the correlation coefficient matrix between the key influencing variables and the secondary influencing variables, and establish the cointegration relationship between the key influencing variables and the secondary influencing variables; use the Jcitest method to test the cointegration relationship between the key influencing variables and the secondary influencing variables , to build a multiple linear regression model. 2. The marine algae growth pollution cointegration prediction method as claimed in claim 1, is characterized in that, the described data related to the growth of marine algae after pretreatment is carried out stability analysis, obtains algae polyvariate time dynamic growth trend curve is specifically: According to the data related to the growth of marine algae after preprocessing, the number of occurrences of marine algae growth pollution or the data related to the growth of marine algae growth showing a decreasing trend is analyzed to obtain a multivariate time dynamic growth trend curve. 3. The marine algae growth pollution cointegration prediction method as claimed in claim 1, is characterized in that, described based on algae multivariate time dynamic growth trend curve, determines the key influence variable and secondary influence variable of algae growth pollution comprising: Based on the multi-time dynamic growth trend curve of algae, the chlorophyll-a concentration of algae is regarded as the key influencing variable of algae growth pollution; Correlation analysis is carried out on the basic data of marine water quality in the data related to the growth of marine algae to obtain secondary influencing variables with high correlation. 4. marine algal growth pollution cointegration prediction method as claimed in claim 3, is characterized in that, also comprises: Delete the secondary influencing variables with low correlation, and use the Dickey Fuller test method to analyze the stationarity of the processed data; The QQ graph analysis method was used to judge whether the core influencing variables conformed to the normal distribution. 5. marine algal growth pollution cointegration prediction method as claimed in claim 1, is characterized in that, described multiple linear regression model is expressed as Y＝1.63X ₁ +4.6516X ₂ +3.6104X ₃ -0.91236X ₄ +1.0207X ₅ -2.563X ₆ +0.52605X ₇ +0.33509X ₈ +0.85703X ₉ -0.023949 Among them, Y is the concentration of chlorophyll a, X ₁ is the five-day biochemical oxygen demand, X ₂ is the content of ammonia nitrogen, X ₃ is the temperature, X ₄ is the salinity, X ₅ is the content of pheomagnesium pigment, X ₆ is the content of silicon, X ₇ is dissolved oxygen content, X ₈ is total nitrogen, X ₉ is total phosphorus content; or, The multiple linear regression model is expressed as Y＝-0.023616X ₁ +1.4527X ₂ -0.29186X ₃ +0.055603X ₄ +0.13686 Among them, Y is the ratio of silicon to phosphorus, _X1 is the average concentration of chlorophyll a in the water column, _X2 is silicate, _X3 is dissolved inorganic nitrogen, and _X4 is the ratio of nitrogen to phosphorus. 6. The marine algae growth pollution cointegration prediction method as claimed in claim 1, is characterized in that, carrying out preprocessing to the relevant data of marine algae growth comprises: Construct a data set related to marine algae growth, and determine whether the missing values in the data set affect the prediction of marine algae growth pollution; If so, the missing value is supplemented by interpolation; if not, the missing value is deleted; Non-numerical data not relevant to marine algae growth pollution predictions were removed. 7. The marine algae growth pollution cointegration prediction method as claimed in claim 1, wherein said marine algae growth-related data include chlorophyll concentration, sea water nutrient concentration and sea water quality data. 8. A cointegration prediction system for marine algae growth pollution, characterized in that it includes: The data processing module is configured to: obtain data related to the growth of marine algae, and preprocess the data related to the growth of marine algae; The co-integration relationship prediction module is configured to: input the pre-processed marine algae growth-related data into the preset co-integration prediction model for processing, and obtain the co-integration relationship between the variables affecting the growth of marine algae, so as to suppress according to the co-integration relationship Marine algae growth pollution; wherein, inputting the preprocessed marine algae growth-related data into a preset co-integration prediction model for processing includes: Perform stability analysis on the preprocessed data related to the growth of marine algae, and obtain the multivariate temporal dynamic growth trend curve of algae; Normalize the data related to the growth of marine algae after preprocessing, and determine the key influencing variables and secondary influencing variables of algae growth pollution based on the multivariate time dynamic growth trend curve of algae; Calculate the correlation coefficient matrix between the key influencing variables and the secondary influencing variables, and establish the cointegration relationship between the key influencing variables and the secondary influencing variables; use the Jcitest method to test the cointegration relationship between the key influencing variables and the secondary influencing variables , to build a multiple linear regression model. 9. An electronic device, characterized in that it comprises a memory, a processor, and computer instructions stored in the memory and run on the processor, when the computer instructions are run by the processor, any one of claims 1-7 can be accomplished described steps. 10. A computer-readable storage medium, characterized in that it is used for storing computer instructions, and when the computer instructions are executed by a processor, the steps described in any one of claims 1-7 are completed.

Images